Exposure of relaxed rat extensor digitorum longus (EDL; predominantly fast-twitch) muscle to temperatures in the upper physiological range for mammalian skeletal muscle (43-46 °C) led to reversible alterations of the contractile activation properties. These properties were studied using the mechanically skinned fibre preparation activated in Ca2+-buffered solutions. The maximum Ca2+-activated force (maximum force per cross-sectional area) and the steepness of force-pCa (-log10[Ca2+]) curves as measured by the Hill coefficient (nH) reversibly decreased by factors of 8 and 2.5, respectively, when the EDL muscle was treated at 43 °C for 30 min and 5 and 2.8, respectively, with treatment at 46 °C for 5 min. Treatment at 47 °C for 5 min produced an even more marked depression in maximum specific force, which fully recovered after treatment, and in the Hill coefficient, which did not recover after treatment. After all temperature treatments there was no change in the level of [Ca2+] at which 50 % maximum force was generated. The temperature-induced depression in force production and steepness of the force-pCa curves were shown to be associated with superoxide (O2−) production in muscle (apparent rate of O2− production at room temperature, 0.055 ± 0.008 nmol min−1 (g wet weight)−1; and following treatment to 46 °C for 5 min, 1.8 ± 0.2 nmol min−1 (g wet weight)−1) because 20 mm Tiron, a membrane-permeant O2− scavenger, was able to markedly suppress the net rate of O2− production and prevent any temperature-induced depression of contractile parameters. The temperature-induced depression in force production of the contractile apparatus could be reversed either by allowing the intact muscle to recover for 3-4 h at room temperature or by treatment of the skinned fibre preparation with dithiothreitol (a potent reducing agent) in the relaxing solution. These results demonstrate that mammalian skeletal muscle has the ability to uncouple force production reversibly from the activator Ca2+ as the temperature increases in the upper physiological range through an increase in O2− production.
Skeletal muscle makes up 40 % of the total body mass in eutherian mammals and, when contracted, can generate a large amount of heat (Rall & Woledge, 1990). Therefore, the temperature of working mammalian muscle can be up to 1-2 °C higher than the body core temperature (Baracos et al. 1984). Under certain conditions, such as high-intensity exercise at high ambient temperature and humidity, the body core temperature approaches 43 °C (Furuyama, 1982). Thus the maximal temperature of working muscle in animals undergoing intense exercise is close to the average temperature threshold (45.3 °C) where irreversible temperature-induced damage of muscle cell integrity occurs (Seese et al. 1998).
Interestingly, there is circumstantial evidence to suggest that the contractile apparatus loses its ability to become activated by Ca2+ and develop force at elevated temperatures. This is illustrated by the lack of any sign of contraction of isolated muscle fibres when the temperature was raised gradually to levels where the integrity of the sarcolemma was compromised and Ca2+ would have entered the sarcoplasm from the external environment (Bischof et al. 1995). The reasons for this loss of contractility occurring at high temperatures have not been previously investigated, although it is known that intact limb muscle preparations display poor longevity at temperatures ≥ 37 °C (Segal & Faulkner, 1985; Lännergren & Westerblad, 1987).
In this study, we used single mechanically skinned rat muscle fibre preparations from dissected whole muscle preparations exposed to elevated temperature and examined the effect of temperatures in the upper physiological and immediate supraphysiological range on contractile apparatus function. The results show that increasing the temperature of skeletal muscle to levels in the extreme physiological range markedly reduces the ability of the contractile apparatus to develop maximum Ca2+-activated force through an increase in the production of the superoxide anion. This depression in force-generating capacity is reversible up to 47 °C.
Animals and muscle dissection
Male Long-Evans hooded rats (16-18 weeks old) were killed by halothane overdose in accordance with the procedure approved by La Trobe University Animal Ethics Committee. The extensor digitorum longus (EDL, a fast-twitch muscle) and peroneus longus (PER, containing predominantly fast-twitch but also a sizeable proportion of slow-twitch fibres; Wang & Kernell, 2000) muscles were rapidly excised, well blotted on filter paper (Whatman no. 1), weighed and then fully immersed in paraffin oil (Ajax Chemicals, Sydney, Australia) or placed in a physiological solution (see below) at room temperature (23 ± 1 °C) contained in a Petri dish.
Isolated EDL muscles were treated to elevated temperatures in a physiological solution containing (mm): NaCl, 145; KCl, 3; CaCl2, 2.5; MgCl2, 1 and Hepes, 10 (pH adjusted to 7.4 with NaOH at the respective temperature), or in paraffin oil that had been heated to a set temperature (within ± 0.1 °C) on a thermostatically controlled hotplate (Model 525, Activon). A magnetic stirrer was placed in the bottom of the beaker to maintain the solution or paraffin oil at a uniform temperature during treatment.
It should be noted that O2 dissolves in paraffin oil, as could be directly demonstrated by measuring the rise in the partial pressure of O2 (PO2) in a solution depleted of O2 that had been covered with a 5 mm layer of paraffin oil using a field O2 probe (YSI 600XL, Yellow Springs International). The PO2 in the vicinity of the resting muscle under oil or in solution was at all times within 15 % of the PO2 in the ambient environment (147 ± 2 mmHg). Considering that PO2 in the muscle in vivo is only 16-22 mmHg (Richmond et al. 1997), it is obvious that under our conditions the muscle was never hypoxic.
Several temperature treatments were performed: 30 min at 43 °C; 5 min at 46 °C; and 5 min at 47 °C. It was important to raise the temperature to a point where at least some of the temperature-induced effects were not fully reversible. This occurred between 46 and 47 °C. After treatment, the muscle was transferred to a Petri dish with paraffin oil at room temperature, where it remained for the entire experimental period after exposure. With this procedure the temperature in the core of the muscle dropped to room temperature within 2 min (n= 3). Single muscle fibres were then isolated and mechanically skinned under paraffin oil (see below) at different times after treatment.
Fibre skinning and mounting
Single muscle fibres were isolated from as close to the surface of the muscle as possible and mechanically skinned under a dissecting microscope using fine forceps (jewelers forceps no. 5) as previously described by Stephenson & Williams (1981). The skinned fibre was then viewed at high magnification (×200) on the screen of a TV monitor coupled to a Leitz dissecting microscope. The fibre could be twisted in such a way as to permit the measurement of fibre width from different positions in at least three places along its length. The cross-sectional area was calculated assuming that the fibre was circular with a diameter equivalent to the average width of the fibre. Note that blotting the muscle dry and keeping it under paraffin oil prevented any gain of fluid, which would have altered fibre diameter. Since no loss of fibre fluid was observed with any of the treatments, it can be safely concluded that the fibre volume at the time of skinning was unchanged compared with that in the muscle immediately after dissection.
The skinned fibre was then attached at one end to a piezoresistive force transducer (AME875, Horten, Norway) using braided silk (Deknatel, size 10 (0.2 mm)) and at the other end to a pair of forceps fixed to a micromanipulator (Fink et al. 1986). With the fibre still immersed in paraffin oil, the resting slack length of the fibre was measured with the aid of a Nikon dissecting microscope at ×10 magnification. The fibre was then transferred to a relaxing solution (Table 1) and the time of immersion into the relaxing solution was noted.
Table 1. Composition of solutions (mm)
Relaxing solution (mM)
Maximum Ca2+ -activating solution (mM)
Sr2+ solution (mM)
The pH of all solutions was 7.10 ± 0.01 at room temperature and the free Mg2+ concentration was 1 mm. Hepes, N-2-hydroxyethylpiperazine-N'-2-ethanesulphonic acid; EGTA, ethyleneglycolbis- (β-aminoethyl-ether) N,N,N',N',-tetraacetic acid; NaN3, sodium azide; ATP, adenosine triphosphate; CP, creatine phosphate.
3 × 10−2
5 × 10−3
The results for various time differences between the beginning of the treatment and the moment of immersion of the skinned fibre into the relaxing solution were pooled together for each temperature treatment in the following groups of time intervals (min): 0-15, 15-30, 30-60, 60-90, 90-120 and 120-150.
Table 1 shows the composition of solutions used in Ca2+-activation experiments. The solutions were prepared as described by Stephenson & Williams (1981). The relaxing solution contained 50 mm EGTA and no added Ca2+ ([Ca2+] < 10−9m) ensuring the complete relaxation of skinned fibres. The maximum Ca2+-activating solution contained almost equimolar amounts of Ca2+ and EGTA as determined by titration (Stephenson & Williams, 1981) and had an ionised [Ca2+] in the order of 3 × 10−5m, sufficient for maximal activation of skinned fibres. The relaxing and activating solutions were mixed in varying ratios to give well-[Ca2+]-buffered solutions with Ca2+ concentrations in the pCa (-log10[Ca2+]) range of ≈4.5 to > 9 (Stephenson & Williams, 1981).
Measurement of Ca2+-activated force
The protocol used for measuring maximum Ca2+-activated force and for obtaining force-pCa curves was as follows. After mounting to the force-recording apparatus, each fibre was placed in the relaxing solution (pCa > 9) for 2 min to equilibrate. The skinned fibre was then placed in the maximum activating solution (pCa ≈4.5) for 5-10 s until the force reached the maximal value and then the preparation was placed back into the relaxing solution for a further 2 min. Force responses were then generated by exposing fibres to activating solutions of progressively lower pCa (higher [Ca2+]) in a stepwise fashion (Fig. 1). There was very little difference (< 10 %) in the ability of the preparation to develop maximum force between the first and the last response in the maximum Ca2+-activated solution and responses at each pCa were expressed as a percentage of the interpolated values for maximum Ca2+-activated force (see Rees & Stephenson, 1987). The experimental data points were best fitted by a Hill equation (eqn (1)) that is characterised by two parameters, Ca50 and the Hill coefficient (nH):
where Ca50 represents the [Ca2+] at which 50 % of maximum Ca2+-activated force is produced and is used as an indicator of the Ca2+ sensitivity of the contractile apparatus and nH, which is proportional to the maximum steepness of the force-pCa curve, refers to the number of Ca2+ ions and cooperativity between the Ca2+-regulatory sites in the functional unit involved in the process of force activation.
Maximum Ca2+-activated force per cross-sectional area (specific maximum Ca2+-activated force, Fmax) was calculated from the magnitude of the maximum force at the beginning of the Ca2+-activation experiments divided by the cross-sectional area measured before the preparation was attached to the force transducer and is expressed in kN m−2.
In order to distinguish between fast- and slow-twitch fibres in the experiments on PER muscle, a strontium (Sr2+)-buffered solution (Table 1) was used. In this solution, Sr2+ was buffered to 5 μm, which is known to maintain fast-twitch fibres in a relaxed state but almost maximally activate slow-twitch fibres (Bortolotto et al. 2000). The fibres were exposed to the Sr2+-buffered solutions at the end of an experiment.
To assess superoxide (O2−) production in the isolated muscle preparation, a combination of the methods described by Kolbeck et al. (1997), Stofan et al. (2000) and Zuo et al. (2000) was employed. The method is based on the reaction of O2− with oxidized cytochrome C (Fe3+) through a one-electron transfer reaction to produce reduced cytochrome C (Fe2+):
When reduced, the absorbance spectrum of cytochrome C has a characteristic peak at 550 nm. Therefore, the relative rise of the 550 nm peak is directly related to the relative amount of the reduced form of cytochrome C in the sample. The relative level of reduced cytochrome C (Cyt C (Fe2+)/Cyt Ctotal) can be calculated by:
where A, B and C are the measured absorbances at 540, 560 and 550 nm, respectively, and D is the value of 2C/(A+B) - 1 when all cytochrome C is in the reduced form. This was achieved by reducing all cytochrome C with O2− produced from the xanthine-xanthine oxidase reaction. Under our conditions, the value of D was 0.64, which is close to the value of 0.7 obtained by Kolbeck et al. (1997).
For O2− measurements, the isolated skeletal muscle was placed in a physiological solution before temperature treatment. Then the muscle preparation was placed in a beaker containing 10 ml physiological solution with 5 μm oxidized cytochrome C (Fe3+) (Sigma) that was heated to 46 °C for 5 min. After treatment, the beaker with the muscle was gradually (15 min) returned to room temperature and the absorbance at 540, 550 and 560 nm was measured in a 2 ml sample of the well-mixed physiological solution using a spectrophotometer (Ultraspec 4050, LKB Biochrom, Cambridge, UK). The cuvette sample was then returned to the bath solution containing the muscle preparation. Measurements were made every 15-30 min. In parallel with the test experiment, a control experiment was also performed, in which another beaker containing 10 ml physiological solution with 5 μm oxidized cytochrome C (Fe3+) without a muscle preparation was also heated to 46 °C for 5 min and was handled in the same way as the beaker that contained the muscle preparation.
Knowing the wet mass of the muscle, M (in grams wet weight (g wet wt)) and the respective absorbance values measured in the samples of physiological solution with (subscript m) and without (subscript c) muscle after a given time, one can calculate the amount of O2− produced by the muscle in nmol O2− (g wet wt)−1 according to the following formula, which was derived from eqn (2) using D= 0.64 and 50 nmol total cytochrome C in the physiological solution:
Results are expressed as means ±s.e.m. Curve fitting and statistical analyses were performed using the scientific analysis program Graphpad Prism (Graphpad Software Inc., San Diego, CA, USA). Statistical significance was tested at P < 0.05 using one-way ANOVA (with Bonferroni multiple comparison test) and two-way ANOVA (with Bonferroni post hoc test), as appropriate.
Temperature effects on specific maximum calcium-activated force (Fmax)
Exposure of the fast-twitch EDL muscle to the upper range of physiological temperatures and the immediate supraphysiological range of temperatures (43 °C for 30 min, 46 °C for 5 min and 47 °C for 5 min), followed by incubation at room temperature (23 ± 1 °C), produced a marked transient reduction in the ability of the contractile apparatus to develop force when maximally activated by Ca2+. This is shown in Fig. 2, where specific maximum Ca2+-activated force (Fmax) values obtained with skinned fibre preparations dissected from the EDL muscle at different time intervals after treatment to 43 and 46 °C are summarised. In Fig. 2 are also shown results from control experiments where muscles were kept at room temperature for the entire period of the experiment. For all temperatures in Fig. 2, the treatment results were highly statistically different from control values (P < 0.001, two-way ANOVA).
After exposure of the EDL muscles to 43 °C for 30 min (Fig. 2A), fibres that were dissected immediately after treatment generated Fmax values that were only 30 % of control values during the same time period. This was followed by a further decline in Fmax to 10 % within 30 min of return to room temperature. After exposure of the EDL muscles to 46 °C for 5 min (Fig. 2B) and 47 °C for 5 min (data not shown) a similar trend was observed. Fibres dissected immediately after treatment generated forces that were about 25 and 75 % smaller than control values (P < 0.001) for the muscles exposed to 46 and 47 °C, respectively. There was then a further loss in Fmax over the following 15-30 min after the muscles were returned to room temperature.
The rate of decline of Fmax when the muscle was exposed to elevated physiological temperatures increased markedly with increasing temperature (6.51 ± 0.28 kN m−2 min−1 after exposure to 43 °C, 12.32 ± 0.36 kN m−2 min−1 after exposure to 46 °C and 13.96 ± 0.35 kN m−2 min−1 after exposure to 47 °C compared with control values of 0.51 ± 0.30 kN m−2 min−1).
Importantly, the ability of the EDL muscle fibres to generate force then showed recovery and over time Fmax returned to values that were not significantly different from control levels. The rate at which Fmax recovered was also dependent on the temperature treatment, with the rate of recovery being markedly lower after treatment at the more elevated temperature. Thus, following exposure to 43 °C for 30 min, fibre Fmax recovered to 80 % of control values within 20 min after reaching a minimum (60 min after exposure). In comparison, after treatment at 46 and 47 °C, Fmax recovered more slowly than after the 43 °C treatment and remained significantly low for about 75 min before returning to control levels at 90-120 min after the 46 °C treatment, while full recovery was not reached until 120-150 min after exposure to 47 °C for 5 min.
The above results were obtained by treating the muscles to elevated temperature in paraffin oil. Experiments where EDL muscles were exposed to 43 °C for 30 min and 46 °C for 5 min in a physiological solution and then transferred to paraffin oil at room temperature produced essentially the same results for each of the two protocols, with Fmax declining to 21.0 ± 0.3 % after 30 min following the 43 °C treatment (n= 5 muscles) and 19.0 ± 0.6 % after the treatment at 46 °C (n= 5 muscles) with full recovery 90-120 min after both treatments. There was no statistical difference between results from experiments conducted in paraffin oil and physiological solution (P > 0.05, two-way ANOVA).
Temperature effects on force-pCa characteristics
Figure 3 shows representative Ca2+-activation curves of individual skinned fibres from a control muscle and from a muscle exposed to 46 °C for 5 min. The force responses were normalised to the maximum Ca2+-activated force response obtained in the same muscle fibres (215 kN m−2 for the control fibre and 29 and 176 kN m−2, respectively, for the fibres dissected 17 and 111 min after exposure of the muscle to 46 °C for 5 min). Muscle treatment at 46 °C also resulted in marked changes in the slope of the force-pCa relationship, which is proportional to nH, but not in the level of Ca2+ that generated 50 % of the maximum Ca2+-activated force (pCa50= -log10[Ca50]). Thus, the pCa50 values for the three fibres shown in Fig. 3 (control fibre, fibre dissected from the 46 °C-treated muscle at a time when the minimum Fmax is reached and fibre dissected from the 46 °C-treated muscle at a time when Fmax has fully recovered) were very close to each other (5.85, 5.92 and 5.91, respectively), while the respective nH values were markedly different (6.39, 1.80 and 4.78, respectively).
There was no statistically significant difference between control and treatment pCa50 values (Fig. 4) for all temperature treatments at 43, 46 and 47 °C and the 43 and 46 °C results are summarised in Fig. 4 and Fig. 5. Interestingly, after muscle exposure to 43 and 46 °C the nH data (Fig. 5) displayed a similar trend to the Fmax data in Fig. 2. Immediately after the 43 °C treatment there was no significant change in nH (6.08 ± 0.04 vs. 6.31 ± 0.64 in controls) but then nH decreased markedly, reaching the lowest level (3.05 ± 0.16) compared with controls (7.62 ± 0.44) 15-30 min after returning to room temperature. Recovery to values that were not significantly different from controls (5.58 ± 0.81 vs. 6.21 ± 0.38) was observed at 135 min after treatment. After treatment at 46 °C for 5 min the lowest value for nH was 2.71 ± 0.63 at 15-30 min after treatment and nH then recovered gradually to levels (5.05 ± 0.19) that were not significantly different from the nH of control fibres (6.2 ± 1.2) at 90-120 min after treatment.
After treatment at 47 °C for 5 min the nH dropped markedly and significantly immediately after the treatment to 2.94 ± 0.21 and there was no measurable recovery of nH for up to 135 min after treatment.
The above results were obtained by treating the muscles to elevated temperature in paraffin oil. When EDL muscles were exposed to 43 °C for 30 min and 46 °C for 5 min in a physiological solution, essentially the same results were obtained for each of the two protocols. Thus nH declined by a factor of 2.7 (n= 5) after 30 min following the 43 °C treatment and a factor of 2.5 (n= 5) after the treatment with 46 °C and there was full recovery after 90-120 min in both treatments. No change in the pCa50 was observed after either temperature treatment. Again there was no statistical difference between results from experiments conducted in paraffin oil and physiological solution (P > 0.05, two-way ANOVA).
Superoxide production after elevated temperature treatment
It has been shown that at temperatures greater than 37 °C (Zuo et al. 2000) reactive oxygen species (ROS), in particular superoxide (O2−), are produced by skeletal muscle. In order to determine whether the effects of elevated temperature treatment on the contractile apparatus could be associated with the generation of O2− it was important to find out whether O2− production did occur under our conditions. For this purpose the reaction of O2− with oxidised cytochrome C (Fe3+) to form reduced cytochrome C (Fe2+) was used (see Methods).
Muscle preparations left at room temperature in a cytochrome C solution had a basal rate of O2− production of 0.055 ± 0.008 nmol min−1 (g wet wt)−1 (Fig. 6). After treatment of the muscles to 46 °C for 5 min, the level of reduced cytochrome C in the test solution containing the muscles increased markedly compared to controls and this is also shown in Fig. 6. During the temperature treatment the apparent maximum rate of superoxide production was 1.8 ± 0.2 nmol min−1 (g wet wt)−1. This rate then slowed down after 60 min to 0.10 ± 0.07 nmol min−1 (g wet wt)−1 for the remainder of the experiment, which is not statistically different from the rate of O2− production at room temperature.
Tiron prevents temperature-induced changes in the contractile properties of EDL muscle fibres
Tiron is a membrane-permeable O2− scavenger that can be used to find out whether the changes in contractile properties observed after temperature treatment are associated with the production of O2− in the muscle. First, it was important to determine at what concentration Tiron was able to suppress cytochrome C reduction in the bathing solution after the EDL muscle was treated for 5 min at 46 °C. In these experiments, whole EDL muscle preparations were pre-incubated for 1 h in a physiological solution with 20 mm Tiron and 10 mm 2,3-butanedione monoxime (BDM). BDM was added to minimise contraction and therefore contraction-dependent generation of O2− (Kolbeck et al. 1997) before muscles were exposed to 46 °C for 5 min under oil and then returned to room temperature. Cytochrome C reduction occurred under these conditions only for the period of the temperature treatment (maximum rate of 1.67 ± 0.19 nmol min−1 (g wet wt)−1 during temperature treatment; Fig. 6) and thereafter the rate of superoxide appearance in the presence of Tiron (0.025 ± 0.020 nmol min−1 (g wet wt)−1) was not significantly different from zero. Therefore, the total amount of cytochrome C reduced (per g wet wt) in the presence of 20 mm Tiron was less than 50 % of the cytochrome C reduced (per g wet wt) in the absence of Tiron. The removal of BDM from the 20 mm Tiron solution (n= 5) did not lead to any increase in O2− production because the results were essentially identical (P > 0.5, two-way ANOVA). Therefore, we can conclude that very little O2− production is contraction-related under our conditions.
Importantly, pretreatment with 20 mm Tiron before exposure to 46 °C for 5 min (Fig. 7) completely prevented any effect on the Fmax for the entire experimental period and markedly reduced the transient depression of nH compared with fibres from muscles not treated with Tiron. Figure 7 also shows the pCa50 of fibres from muscles that were incubated in 20 mm Tiron and, as expected, no change was observed. EDL muscles (n= 3) were also incubated in 40 mm Tiron to find out whether the effect on nH could be further minimised compared with 20 mm Tiron. However, no further improvement was observed in these experiments (P > 0.05; data not shown).
In conclusion, the results with Tiron unequivocally show that the temperature effects on Fmax and nH must be predominantly due to the generation of O2− in the muscle induced by higher temperatures.
Tiron has differential protective effects on fast-twitch and slow-twitch fibres
It was also of interest to use a mixed muscle containing a sizable proportion of slow- and fast-twitch fibres to see if the two types of fibres are similarly affected by the same temperature treatment. Peroneus longus (PER) muscle is such a mixed muscle (Wang & Kernell, 2000) and an experiment was performed with a PER muscle in a physiological solution incubated in 20 mm Tiron for 1 h and then exposed to 46 °C for 5 min under paraffin oil and returned to room temperature under paraffin oil. From this muscle, fibres were dissected, identified as fast- or slow-twitch using the Sr2+-activation test (see Methods) and then maximally activated by Ca2+. The results obtained from six fibres are shown in Fig. 8. The fast-twitch fibres produced forces that were not significantly different from controls or from fibres taken from EDL muscles treated with Tiron, as expected. However, the Fmax in all slow-twitch fibres from the same muscle was markedly depressed and significantly lower than that in the fast-twitch fibres despite the fact that Fmax should be similar in both fibre types under control conditions (Stephenson & Williams, 1981).
Effect of dithiothreitol (DTT) on the contractile properties after temperature treatment
As indicated earlier, O2− can act as a reducing agent in some reactions, such as the reduction of cytochrome C, but more often it acts as an oxidising agent (as in the oxidation of ascorbic acid shown below) leading to the production of ROS, such as H2O2:
In order to distinguish between the type of O2− reaction that is mainly responsible for the O2−-induced effects on the EDL muscle, some experiments were repeated in the presence of DTT, which is a membrane-permeant, powerful reducing agent. If a reducing reaction alone were responsible for the marked changes in force characteristics, then one would expect that treatment of the muscle with DTT would induce similar effects because the temperature treatment and, importantly, pretreatment of the muscle with DTT would reduce or completely prevent changes in Fmax and nH observed following temperature treatment.
In one experiment, fibres were exposed to 20 mm DTT for 10 min without temperature treatment, which had no effect on any of the parameters measured (Fmax, nH and pCa50; data not shown). In another experiment, isolated EDL muscles were initially pre-incubated in a physiological solution containing 20 mm DTT for 1 h. Thereafter, the muscles were subjected to treatment at 46 °C for 5 min in paraffin oil and fibres were dissected and mechanically skinned after the muscle was returned to room temperature. This protocol enabled us to see whether the presence of DTT in the muscle before temperature treatment caused a reduction in the temperature-induced loss in force observed in the absence of DTT. The changes in Fmax and nH induced by temperature treatment in the absence of DTT were not prevented by the presence of DTT. In fact, Fmax for fibres from DTT-treated muscles decreased significantly more (by a factor of 8 compared to a factor of 5) than for fibres from muscles exposed to elevated temperature without DTT and also recovered less after return to room temperature (40 compared to 85 % recovery, one-way ANOVA, P < 0.05). Similar results were seen in the nH changes, where there was a decrease followed by only a partial recovery, smaller than for fibres from muscles not incubated in DTT. As expected, the pCa50 was not affected at all.
These results demonstrate that the O2−-induced effects on Fmax and nH are more marked in the presence of a powerful reducing agent, suggesting that the O2−-induced effects on Fmax and nH are most probably due to its oxidising properties. This could be because O2− is involved in generation of other ROS or in oxidising certain groups on the contractile apparatus and regulatory system that are even more sensitive to oxidation in a strongly reducing environment.
In order to further characterise mechanisms responsible for the effects on Fmax and nH, additional experiments were performed with skinned fibres, where DTT was applied directly on the contractile apparatus under controlled conditions after the muscle was exposed to 46 °C for 5 min. The skinned fibres were exposed to DTT for 10 min in a relaxing solution either before or after being maximally activated by Ca2+. As shown in Fig. 9, if the fibres were initially maximally activated before treatment with DTT, the Fmax did not change, remaining at the same level as it was before DTT treatment. However, if skinned fibres were dissected from high temperature-treated EDL muscles and were then incubated in the relaxing solution with 20 mm DTT for 10 min before any activation (Fig. 9), the Fmax produced was not significantly different from control values when the muscle was not subjected to the temperature treatment. Even though DTT treatment before any activation prevented the drop in Fmax, it could not prevent the temperature-induced decrease in nH. As expected, there was no change in the pCa50.
Overall, the DTT results suggest that: (i) the temperature (O2−)-induced effects on the ability of the contractile apparatus to develop Fmax are due to a direct or indirect (via other ROS) oxidising action by O2− because DTT can reverse the action of O2−; (ii) the temperature (O2−)-induced Fmax depression is dependent on the activation of the contractile apparatus because once the contractile apparatus was activated, DTT could not reverse the effects; and (iii) the temperature (O2−)-induced effects on the nH are due to the oxidising action of O2− on different groups from those responsible for the depression in Fmax.
Temperature effects on the contractile apparatus
The present results show that exposure of mammalian skeletal muscle to temperatures in the range expected to occur under extreme physiological conditions have a significant effect on the ability of the contractile apparatus to develop force when maximally activated by Ca2+. Increasing the temperature from 23 ± 1 to 43, 46 and 47 °C increased the rate at which the contractile apparatus in fast-twitch fibres lost the ability to produce force from 0.024 ± 0.01 to 2.36 ± 0.02, 4.98 ± 0.03 and 16.00 ± 0.03 % Fmax min−1, respectively (Fig. 2). This was paralleled by a marked decrease in the steepness of the force-pCa curves without change in the [Ca2+] corresponding to 50 % maximum Ca2+-activated force (Fig. 5). The rates of loss and recovery of contractile parameters at 46 and 47 °C would be greater considering that muscles were only briefly exposed to these temperatures (5 min). It is important to note that the muscles were not hypoxic under any conditions used in this study (see Methods).
Importantly, the ability to generate force and the depression of the Hill coefficient then showed recovery and gradually returned to values that were not significantly different from control levels. The rate of recovery was also shown to be dependent on the temperature treatment, with the rate of recovery decreasing after treatment at the more elevated temperatures. This phenomenon is linked to the production of O2− in muscle and has important physiological significance.
The basal rate of O2− production in EDL muscle at room temperature was 0.055 ± 0.008 nmol min−1 (g wet wt)−1. During treatment at 46 °C for 5 min the rate of O2− production increased at least 30-fold considering that our measured rate of 1.8 ± 0.2 nmol min−1 (g wet wt)−1 over the first 5 min must have been limited by diffusion. This result is similar to that obtained by Zuo et al. (2000), who found that when rat diaphragm muscle was at 42 °C for 45 min, the rate of O2− production was about 1.0 nmol min−1 (g wet wt)−1. Here we have shown that the transient decrease in the ability to develop force was directly related to O2− production because the effects induced on the contractile activation characteristics by elevated temperature treatment could be minimised by the presence of a potent, membrane-permeable O2− scavenger (20 mm Tiron; Fig. 7).
There are a number of potential sources of O2− production in skeletal muscle. Some of these sources are associated with the muscle fibres while others are external to muscle fibres. For example, O2− has been shown to be produced by the endothelial cells of blood vessels in the muscles (Hellsten-Westing, 1993) and also via membrane-bound NADPH- and/or NADH-oxidases (De Keulennaer et al. 1998), although the major source of O2− in the muscle is probably the mitochondria (Zuo et al. 2000).
Results with the PER muscle (Fig. 8) incubated in a physiological solution with 20 mm Tiron suggest that, at this concentration, Tiron was sufficient to scavenge the O2− produced in the fast-twitch fibres but not sufficient to scavenge the O2− produced in the slow-twitch fibres treated to 46 °C for 5 min before O2− acted on the contractile apparatus. These results also indicate that, under our conditions, O2− is mainly produced within the muscle fibres themselves rather than in the cells and structures extraneous to the muscle fibres, such as capillaries. This is because if O2− was mainly produced external to the muscle fibres, then both fast- and slow-twitch fibres should have been equally affected by O2−, assuming similar sensitivity to O2−. The results can be explained if mitochondria are the major source of O2− production in muscle because the slow-twitch fibres, which appeared to be more affected by the temperature treatment, are likely to contain a higher density of mitochondria than fast-twitch fibres (Eisenberg, 1983).
Mechanism of O2− action on the contractile apparatus
Our results unequivocally showed that the depression in the force production and Hill coefficient are directly associated with the production of superoxide. It is most likely that the effects observed are due primarily to the effects of O2− rather than its byproducts H2O2 and OH•. Indeed, other studies in which O2− was directly generated in solution have shown effects on the contractile apparatus that are similar to those found in this study. Thus, MacFarlane & Miller (1992), Lawler et al. (1998) and Darnley et al. (2001) have shown that exposure of myofilaments to O2− caused force production to decline and that there was no significant change in sensitivity to Ca2+ (pCa50); results which are consistent with our observations on the contractile properties after exposure to elevated temperatures.
From our results, we have evidence suggesting that O2− acts mainly as an oxidising agent on the contractile apparatus. Thus, when DTT was added to mechanically skinned fibres from muscles treated to elevated temperature, the resulting Fmax was not significantly different from that of controls. Since DTT is a strong reducing agent and is sulfhydryl group specific, we can infer that the temperature-induced loss in force generation is due to a O2− oxidation reaction on the SH groups on the contractile apparatus and not to a O2− reducing reaction such as that with cytochrome C.
Most elements of the contractile apparatus and Ca2+ regulatory system, such as myosin, actin, tropomyosin and troponin, contain SH groups (Wilson et al. 1991). The oxidation of one or more of these protein structures could disrupt the interactions between them. The most sensitive SH groups are located on the myosin filaments. Myosin has been shown to contain two SH groups per head which, when covalently modified with oxidising agents, cause a reduction in maximally Ca2+-activated force and changes in the force-pCa relationship which are reversible with reducing agents (Wilson et al. 1991). Whilst this explains the temperature-induced changes in the maximum force response, it is noteworthy that DTT failed to recover the Hill coefficient. If the changes observed in nH (Fig. 6) were also due to oxidation of the same SH groups, then DTT would have been expected to reverse the temperature-induced effects. Since this was not the case, the changes in nH must be due to oxidation/reduction reactions of groups on the myofilament proteins other than SH groups that, when altered, affect Fmax.
There is a wide range of antioxidants and antioxidant enzymes scattered throughout the muscle cell that are important in combating the effects of O2− on muscle function, such as superoxide dismutase, vitamin E and vitamin C. The presence of antioxidants reduces the lifetime of O2− and its byproducts, thus limiting their effects on muscle function, as shown by the results with Tiron (which acts like an antioxidant), where the temperature effects on the Fmax were completely prevented. The fact that the depression in the force response and in the steepness of the force-pCa curves are reversed when the muscle is returned to room temperature indicates that reducing agents present in the muscle fibres are responsible for this recovery. Without these reducing agents, the observed temperature effects would be irreversible. Interestingly, when muscles were incubated in a highly reducing environment with 20 mm DTT present before temperature treatment there was an even stronger O2− effect on the contractile apparatus. This could be explained by the cellular redox balance of the muscle cell being pushed further into the reduced state by DTT, where the reactivity to O2− and its byproducts of the various groups on the contractile apparatus proteins increases (Andrade et al. 1998).
As far as we are aware, this study is the first to combine the production of superoxide at elevated temperatures and muscle function. Our results suggest that the reason for this loss of contractility at temperatures in the upper physiological range is due to an increase in O2− production in skeletal muscle as muscle temperature rises. Therefore, if the temperature of the working muscle rises in the upper physiological range, the production of O2− increases and the contractile apparatus loses its ability to contract and to generate heat, thus preventing the body core temperature from rising above the lethal temperature if the O2− is not rapidly removed. Of course, O2− could also act on sites associated with cellular membranes, such as sarcoplasmic reticulum (Brotto & Nosek, 1996), which would affect their function. As the temperature of the muscle decreases, the endogenous reducing agents would reverse these temperature-induced effects and the contractile apparatus would regain its ability to produce force. This may act as a protective mechanism that stops body core temperature from reaching the lethal level. Importantly, displacing the redox potential of the muscle towards a more reducing state can be even more effective in uncoupling force development and heat production from the activator Ca2+ as the temperature rises. This process of decreasing force production may provide eutherian mammals with a built-in biological ‘switch’, based on oxidation/reduction effects, that prevents muscle contraction at high temperature, thereby avoiding body temperature increases that could result in muscle damage and even animal death.
We thank the National Health and Medical Research Council of Australia for financial support, Mr P. Wiggins for help with PO2 measurements and Drs G. D. Lamb, B. S. Launikonis and N. Ortenblad for discussion.